Research On The System Of Prognostics And Health Management Based On Special Vehicle

The special vehicle in this paper is an important equipment in the aerospace field,and its reliability is very important for the completion of the mission.Due to the frequent use and the complexity of equipment,special vehicles often lose efficacy in the implementation of the task,affecting the normal task.With the integrated,comprehensive and intelligent level of special vehicle equipment higher,the risk of equipment development is more and more bigger,period is more and more longer,costs is more and more higher.At the same time,monitoring of the running state of the equipment and maintenance methods are also put forward higher requirements.In order to realize the comprehensive health management of special vehicles,based on the study of analysis methods of fault tree,expert system and BP neural network,the paper put forward a new method of fault diagnosis,which is the combination of fault tree and rule,and established the BP neural network prediction model for special vehicles to forecast the possible fault of special vehicle.Firstly,based on the analysis of the actual needs of the special vehicle,the paper designed the system architecture and software framework of special vehicle prognostics and health management system,and realized the import of test data and related document configuration through the analysis of the original test data of vehicles.Then,the paper built the special vehicle fault tree model based on the analysis of special vehicle fault mode,and through the configuration of fault tree node attribute,the paper realized graphics rendering and display function based on CStatic control.Finally,based on the analysis of special vehicle history running data,the paper set up the special vehicle health grade samples,which were used to train and optimize the BP neural network prediction model of special vehicle,and the trained BP neural network prediction model was solidified and embedded to software platform.The innovation of this paper mainly has two aspects.The first is the universality of special vehicle prognostics and health management software platform.Through the import of a series of configuration files,such as algorithms configuration files,rule configuration files,and fault tree configuration files,software platform does not rely on any one kind of specific special vehicle.As long as having the data files and the associated configuration files,it can do the work of health management of any series of special vehicles.The second is the graphic display method of fault tree based on CStatic control.Based on the combination of node properties of the fault tree CStatic control,the system has realized thevisualization and diversification of fault tree graphic.Through this method,the system can draw fault tree automatically and users can browse the fault tree according to their own needs.The prognostics and health management system is verified by the historical running data of the special vehicle.The diagnosis and prediction results of the software show the versatility and practicability of the system,being up to the actual mustard.